The Financial Crisis of 2008: What Went Wrong? David Marshall * Senior Vice President S i Vi P id t Federal Reserve Bank of Chicago 27 th Annual Options Industry Conference 27 Annual Options Industry Conference Weston, Florida May 1, 2009 * The opinions in this presentation are the presenter’s and do not reflect positions of the Federal Reserve Bank of Chicago or the Federal Reserve System . 1
Prices and Value � A self-correcting economic system: – People respond to price incentives – Prices (approximately) reflect fundamental value Prices (approximately) reflect fundamental value Leading into the current crisis, people responded to prices � – Yields on risky securities fell, so producers of these securities ramped up production. – Prices of residential real estate were high and rising, so homebuilders built more homes. – Real long term interest rates fell, so households reduced savings and increased consumption. g p � But prices did not reflect fundamental value: – Price of risk too low Price of risk too low – Price of residential real estate too high 2
Plan of Talk � Fundamental driver of the crisis � Mi � Mispricing # 1: Risk i i # 1 Ri k � Mispricing # 2: Housing Mispricing # 2: Housing � The two mispricings come together: Subprime mortgage securitization t iti ti � The role of the Fed The role of the Fed � Policy conclusions going forward 3
The Fundamental Driver � Massive global capital flows into the U.S. 1998 - 2006 – Increased productivity of East and South Asian workers Increased productivity of East and South Asian workers – High saving rate – Savings invested in U.S. financial markets g – Anomaly: Poor countries sending capital to rich. � Result: very low real interest rates R lt l l i t t t 4
Flow of Capital into U.S. and Real Interest Rates 7 7 6 6 ) Term Rates (%) 5 Capital (%) 5 4 3 4 Real Long T Flow of 2 3 1 2 0 0 ‐ 1 1 Flow of Capital into U.S. as % of GDP Real 30yr Mort. Rates 5
Price of Risk: VIX + Junk Bond Spread 70 7 60 60 6 6 Spread (%) 50 5 x Value 40 4 VIX Index Junk Bond S 30 3 20 2 10 1 0 0 VIX Junk Bond Spread 6
Low Risk Pricing � Typical explanation: “The great moderation” – Business cycle variability since 1984 dramatically lower Business cycle variability since 1984 dramatically lower – � Lower fundamental risk in the economy 7
Lower Risk or Higher Risk Tolerance? � Alternative explanation: Not only a reduction in risk, but also an increase in risk tolerance. � Strong incentives for portfolio managers to seek out risk: – Low real interest rates lead investors to search for yield. L l i t t t l d i t t h f i ld – Only way to increase yield: take on additional risk. – Portfolio manager compensation contracts provided further – Portfolio manager compensation contracts provided further incentives for risk-taking � Bonuses for short-term performance � Hedge fund compensation based on “high water mark” 8
Tail Risk � Managers would preferentially take on tail risk: – Risk of low probability but high impact events. – Example: risk of a broad-based house price decline. � Tail risk is extremely difficult to quantify � Tail risk is extremely difficult to quantify. � Most risk management approaches measure risk by short-run volatility, which can’t capture tail events. h t l tilit hi h ’t t t il t � High yields associated with tail-risk strategies show up as α (high risk-adjusted performance) while it really represents β (compensation for risk). � AIG Financial Products 9
Mispricing of Housing � Surplus capital from abroad increases demand for U.S. securities. – This demand was met by securitizing residential real estate MBS securitizers needed flow of mortgages to satisfy demand � – Vast expansion of sub-prime mortgage origination p p g g g � Major public policy push to extend home-ownership to previously underserved households. p y � Volume of sub-prime mortgages soared to meet this demand – From 2000-2007, From 2000 2007, � Outstanding amount of conforming mortgages doubled, � But subprime grew 800%! – By 2006 , Subprime /Alt-A mortgage issuance ≈ 30% of the mortgage market 10
Housing Prices 1987-Present 200 180 160 140 ex Value 120 100 Inde 80 60 40 20 0 Case ‐ Shiller Index 11
Housing Starts: 1960 - Present 3000 2500 tarts 2000 ousands of S 1500 Tho 1000 1000 500 0 12
Housing Bubble � How did housing prices get so far above fundamental value? – It’s hard to detect a bubble when you’re in the middle of it. – Housing starts in 2000 – 2005 within historical range. – Analysis in 2006: increased housing investment could be justified by fundamentals be justified by fundamentals � Increased household wealth � Financial innovation � Financial innovation. 13
Subprime Mortgage Securitization � Demand for tail risk + housing bubble combined to create a perfect storm: Subprime MBSs � Design of MBSs: equity tranche protects the senior tranches. – Allows senior tranches to achieve AAA status – But: Equity tranche only provides protection if defaults within the mortgage pool have low correlation. – Rating agencies estimated these correlations from past data. g g p – Impact of tail events on these correlations not taken into consideration 14
Subprime Mortgage Securitization � Problem: Sub-prime mortgages much more sensitive to house price decline than conventional mortgages – When housing prices fall (as in 2006), all the correlations Wh h i i f ll ( i 2006) ll th l ti become extremely high! � Senior tranches unprotected, fall from AAA to junk. S i t h t t d f ll f AAA t j k � These formerly AAA-rated assets are now the “toxic assets” – Clogging bank balance sheets – Limiting liquidity provision – Starving the economy for credit – Starving the economy for credit 15
Did the Fed play a role in the housing bubble? � A story put forth by some observers: – The housing bubble was caused by excessively easy The housing bubble was caused by excessively easy monetary policy from 2001 – 2004. � To evaluate this story, we need to ask: T l t thi t d t k – Was the Fed’s monetary policy excessively easy given the circumstances prevailing in 2001 – 2004? given the circumstances prevailing in 2001 2004? – How much could the housing bubble have been offset by a more restrictive monetary policy from 2001 – 2004? 16
Federal Funds Rate 2000 - 2007 7 6 Excessive Liquidity? 5 4 cent Per 3 2 1 0 Effective Fed Funds Taylor Rule 17
Monetary policy in 2001 - 2004 � The Fed's legal mandate is to foster: – Maximum sustainable employment – Price stability � Employment in 2001 � Employment in 2001 – 2003 2003 – The 2001 recession ended in November, but the economy continued shedding jobs until Q2:2003! – Fed continued cutting rates well into 2003. � Price stability in 2001 – 2003 � Price stability in 2001 – 2003 – Inflation rate measured in real time fell to 0.7% in June 2003. – Flirting with deflation, much more debilitating than inflation. – Fed response: continue to cut interest rates to 1%. 18
Could the Fed have offset the housing bubble? � Fed controls the very short interest rate � Housing responds mainly to long term interest rates � During this period the link between long and short � During this period, the link between long and short rates was unusually weak. – Likely reason: huge capital inflow from abroad Likely reason: huge capital inflow from abroad pushing long rates down. 19
Relationship between long rates and short rates 25 20 15 cent Perc 10 5 0 10 ‐ Year T ‐ Note Yield 10 Year T Note Yield Fed Funds (Effective) Fed Funds (Effective) 20
Fed’s Response to the Crisis: Provide Liquidity � Vast expansion of Fed balance sheet – August 2007: $900 billion August 2007: $900 billion – April 2009: $2 trillion � Innovative programs to inject liquidity into specific markets – Term Auction Facility (TAF) T A ti F ilit (TAF) – Liquidity Swaps with Foreign Central Banks – Commercial Paper Funding Facility (CPFF) Commercial Paper Funding Facility (CPFF) – Agency MBS purchase program – Term Asset-Backed Securities Loan Facility (TALF) Term Asset Backed Securities Loan Facility (TALF) 21
22
Long Run Policy Implications � Measurement and containment of tail risk – Managerial incentives – Disruptive effects of housing mispricing � Thorough revision of failure resolution procedures � Thorough revision of failure resolution procedures – Bank failure resolution through FDIC works fairly well – No comparable procedures for bank holding companies or non- banks � Should the Fed pop asset price bubbles? Should the Fed pop asset price bubbles? – Can the Fed identify bubbles in a timely fashion? – Can Fed action have a substantial impact on a bubble? – Will the benefits from popping a bubble exceed the cost? 23
Recommend
More recommend
